HAYAKAWA, Yu

写真a

Affiliation

Faculty of International Research and Education, School of International Liberal Studies

Job title

Professor
To the head of this page.▲

Concurrent Post 【 display / non-display 】

  • Faculty of Science and Engineering   Graduate School of Fundamental Science and Engineering

To the head of this page.▲

Education 【 display / non-display 】

  • 1986
    -
    1992

    University of California at Berkeley   College of Engineering   Operations Research  

  • 　
    -
    1986

    Illinois State University   Department of Mathematics   Mathematics  

  • 　
    -
    1983

    Hiroshima University   Faculty of School Education   Elementary School Education  

  • 　
    -
    1983

    Hiroshima University   Faculty of School Education   Elementary School Education  

To the head of this page.▲

Degree 【 display / non-display 】

  • University of California at Berkeley   Doctor of Philosophy

To the head of this page.▲

Research Experience 【 display / non-display 】

  • 2007
    -
    　

    Professor, Waseda University (1 May Present)

  • 2004
    -
    　

    Associate Professor, Waseda University (1 April April 2007)

  • 2001
    -
    　

    Senior Lecturer, Victoria University of Wellington, New Zealand (1 January March 2004)

  • 1992
    -
    　

    Lecturer, Victoria University of Wellington, New Zealand (1 September December 2000)

To the head of this page.▲

Professional Memberships 【 display / non-display 】

  • 　
    　
    　

    IEEE

  • 　
    　
    　

    Reliability Engineering Association of Japan

  • 　
    　
    　

    Japanese Society of Applied Statistics

  • 　
    　
    　

    Japan Statistical Society

  • 　
    　
    　

    Operations Research Society of Japan

To the head of this page.▲
 

Research Interests 【 display / non-display 】

  • Operations Research

To the head of this page.▲

Papers 【 display / non-display 】

  • Discussion of “Virtual age, is it real?”

    Richard Arnold, Stefanka Chukova, Yu Hayakawa

    Applied Stochastic Models in Business and Industry    2020  [Refereed]  [Invited]

  • Mean and Variance of an Alternating Geometric Process

    Richard Arnold, Stefanka Chukova, Yu Hayakawa, Sarah Marshall

    2020 Asia-Pacific International Symposium on Advanced Reliability and Maintenance Modeling (APARM)    2020  [Refereed]

    DOI

  • Nonparametric Bayesian Analysis of Hazard Rate Functions using the Gamma Process Prior

    Richard Arnold, Stefanka Chukova, Yu Hayakawa

    2020 Asia-Pacific International Symposium on Advanced Reliability and Maintenance Modeling (APARM)    2020  [Refereed]

    DOI

  • Delayed Reporting of Faults in Warranty Claims

    Richard Arnold, Stefanka Chukova, Yu Hayakawa

    IEEE Transactions on Reliability   69 ( 4 ) 1178 - 1194  2020  [Refereed]

    DOI

  • Nonzero repair times dependent on the failure hazard

    Richard Arnold, Stefanka Chukova, Yu Hayakawa, Sarah Marshall

    Quality and Reliability Engineering International   36 ( 3 ) 988 - 1004  2020  [Refereed]

display all >>

To the head of this page.▲

Research Projects 【 display / non-display 】

  • Non-parametric Bayesian approach to modelling system reliability

    Project Year :

    2018.04
    -
    2021.03
     

  • Non-parametric Bayesian approach to modelling system reliability

    Grants-in-Aids for Scientific Research (C)

    Project Year :

    2018.04
    -
    2020.03
     

  • New models for failures in multi-component systems: Model properties and simulation methods

    Project Year :

    2013.11
    -
    2014.02
     

  • Robustness measures for telecomunication networks

    Project Year :

    1993
    -
    1994
     

    　View Summary

    jointly with Peter J Smith

To the head of this page.▲

Presentations 【 display / non-display 】

  • Schur-concavity, aging and a new model for damage accumulation in multi-component systems (Arnold, R., Chukova, S. and Hayakawa, Y.)

    3rd Asian International Workshop on Advanced Reliability Modeling (AIWARM 2008) 

    Presentation date： 2008.10

  • Capture Recapture estimation using finite mixtures of arbitrary dimension (Arnold, R. and Hayakawa, Y.) presented by Arnold, R.

    9th World Conference of the International Society for Bayesian Analysis (ISBA) 

    Presentation date： 2008.07

To the head of this page.▲

Specific Research 【 display / non-display 】

  • Non-parametric Bayesian approach to modelling system reliability

    2018   Richard Arnold, Stefanka Chukova, Sarah Marshall

    　View Summary

    Weare taking a nonparametric Bayesian approach to hazard function modelling. Thehigh level of complexity of this approach has meant we spent a significantamount of time on review of the literature and on model formulation.  We have created a draft technical reportsummarising our findings and have implanted simulation models for the GammaProcess methodology which we are using. We are also working on biclustering incapture-recapture experiments from a nonparametric Bayesian perspective and arepresenting our results at MMR 2019.  Wewrote a paper on delayed reporting of faults in warranty claims, in which thereporting process is modelled as a stochastic process dependent on theunderlying stochastic process generating the faults.  Our paper has been presented at seminars andconferences. This research inspired another project on nonzero repair timesdependent on the failure hazard.  Ourpaper on this project will be presented at MMR 2019. Wealso worked on a review paper and a book chapter on geometric-like processesand their applications in warranty analysis. Conference and workshop presentations have been given on this topic andwe also present our paper at MMR 2019. The book chapter has been accepted for publication.

  • Warranty cost analysis with non-zero repair times through hierarchical stochastic processes

    2017   Richard Arnold, Stefanka Chukova, Yuuki Rikimaru

    　View Summary

    1)    We model the warranty claims process and evaluate the warranty servicing costs under non-renewing and renewing free repair warranties.  We assume that the repair time for rectifying the claims is non-zero and the repair cost is a function of the length of the repair time.  To accommodate the ageing of the product and repair equipment, we use a decreasing geometric process to model the consecutive operational times and an increasing geometric process to model the consecutive repair times.  We identify and study the alternating geometric process (AGP), which is an alternating process with cycles consisting of the item’s operational time followed by the corresponding repair time.  We derive new results for the AGP in finite horizon and use them to evaluate the warranty costs over the warranty period and over the life cycle of the product under a non-renewing free repair warranty (NRFRW), a renewing free repair warranty (RFRW) and a restricted renewing free repair warranty (RRFRW(n)).  Properties of the model are demonstrated using a simulation study.Together with Dr Sarah Marshall, Auckland University of Technology, we wrote a paper on the results of this project and it was made a technical report (Research Report Series (ISSN:1174-2011), #18-1, Victoria University of Wellington School of Mathematics and Statistics (http://sms.victoria.ac.nz/Main/ResearchReportSeries) (2018)).  We submitted this paper to a joint conference: APARM2018 & QR2MSE2018.  We will present this paper at this conference in August 2018.2)    We present a model for the delayed reporting of faults: multiple non-fatal faults are accumulated and then simultaneously reported and repaired.  The reporting process is modelled as a stochastic process dependent on the underlying stochastic process generating the faults.  We have worked on the cases of multiple faults types, planned preventative maintenance and customer rush.  A manuscript has been written on the results and we plan to extend it by adding simulation studies and submit it to an international peer reviewed journal this year.

  • Delayed reporting of faults in warranty claims

    2017   Richard Arnold, Stefanka Chukova

    　View Summary

    When a complex system is operated, it may experience multiple faults. If the system is operating under warranty these faults may be claimed for and repaired at zero or minimal cost to the consumer.  However, if the faults do not lead to system failure the user may find it inconvenient to claim for each repair as it occurs and may instead delay making a report or claim until a sufficiently large number of faults has accumulated. We present a model for the delayed reporting of faults: multiple non-fatal faults are accumulated and then simultaneously reported and repaired. The reporting process is modelled as a stochastic process dependent on the underlying stochastic process generating the faults. The joint distribution of the reporting times and numbers of reported faults is derived.  We presented a paper on the project in the 10th International Conference on Mathematical Methods in Reliability held in France on 3-6 July 2017.  We also worked on a few extensions of the above model, which deal with multiple fault types, planned preventative maintenance and customer rush and wrote a manuscript on the results.  We have written a manuscript on the above results and plan to extend it by adding simulation results.  An extended version of the manuscript will be submitted to an international peer review journal later this year.

  • Modelling system reliability: Non-parametric Bayesian inference

    2016   Richard Arnold, Stefanka Chukova, Adam Rod

    　View Summary

    1) We construct a model for the delayed reporting of non-fatal faults which are accumulated before being reported in groups, which is applicable to warranty claim analysis. We extend this model to account for multiple fault types, planned preventative maintenance and customer rush. We have submitted a manuscript for presentation in an invited session at the 10th International Conference on Mathematical Methods in Reliability in Grenoble, France, on 3-6 July 2017. We also presented our paper at 2016 Joint NZSA+ORSNZ Conference (27-30 November 2016) and the Joint Workshop – New Development of Stochastic Models (15 October 2016). Our paper on this model was awarded a Best Paper award at the Asia-Pacific Workshop on Advanced Reliability and Maintenance Modelling (Seoul, August 2016).2) We present a general approach to inference in multicomponent systems where the system is viewed as being made up of Independent Overlapping Subsystems (IOS). We submitted a paper on this project to IEEE Transactions on Reliability in February 2016 and submitted a revision of the paper in December 2016.3) We present a unified model for multi-component systems with dependent failures and imperfect repairs. A paper on this project was finalised and submitted to “Stochastic Operations Research in Business and Industry,” World Scientific, for inclusion as a book chapter in December 2016.

  • Stochastic process approaches to modelling system reliability

    2015   Richard Arnold, Stefanka Chukova, Fan Wang

    　View Summary

    1) We present a model for the delayed reporting of faults: multiple sub-critical faults are accumulated and then simultaneously reported and repaired. The reporting process is modelled as a stochastic process dependent on the underlying stochastic process generating the faults. The joint distribution of the reporting times and numbers of reported faults is derived.  We have submitted a manuscript for presentation at The 7th Asia-Pacific International Symposium on Advanced Reliability and Maintenance Modeling in Seoul, Korea, on 24-26 August 2016.2) We generalise existing approaches to inference in multicomponent systems by formulating the likelihood for systems made up of independent overlapping subsystems, and have shown how to apply this inference procedure to the models that we had previously published.   A paper on this project was finalised and submitted to IEEE Transaction on Reliability in February 2016.3) We model the warranty servicing costs by assuming non-zero increasing repair times.  We introduce the generalised alternating renewal process by using the geometric process.  Simulation is used to estimate the expected costs over the warranty period and life cycle.  We have written a manuscript summarising our results and plan to submit its finalised version to European Journal of Operational Research by August 2016.

display all >>

To the head of this page.▲

Overseas Activities 【 display / non-display 】

  • Modelling system reliability via stochastic process

    2016.09
    -
    2017.03

    New Zealand   Victoria University of Wellington

  • Studies on damage accumulation models

    2011.09
    -
    2012.03

    ニュージーランド   Victoria University of Wellington

To the head of this page.▲
 

Syllabus 【 display / non-display 】

display all >>

To the head of this page.▲
 

Committee Memberships 【 display / non-display 】

  • 2020.06
    -
    　

    Reliability Engineering Association of Japan  Vice-president

  • 2020.06
    -
    　

    日本信頼性学会  副会長

  • 2015.01
    -
    2016.12

    IEEE Reliability Society Japan Joint Chapter  Chair

  • 2015.01
    -
    2016.12

    IEEE Reliability Society Japan Joint Chapter  チェア

To the head of this page.▲